My Neighbor Etchant Lab, or: titanium Pebbles

@astosia and I have been collaborating for probably a year now on various different housings for Pebbles (mostly Pebble 2 form-factor compatible watches – so Pebble 2, and Core 2 Duo). She’s been doing a bunch of amazing modeling of watches, and I’ve been experimenting with metal manufacturing processes – in particular, getting things made out of SLM (an additive 3D printing technology that results in fully dense metal) titanium.

Titanium is a cool material because it’s pretty darn sturdy, and you can get a bunch of interesting surface finishes on it. I’ve been posting occasionally on Discord, but Discord kind of sucks, so I wanted to create a thread here to track what I’ve been working on.

The first run of DMLS Ti watches I did, I did with ProtoTi, back in May last year. The raw parts had a bead-blasted finish; I didn’t try to clean them up, and anodized a few of them blue. I FDM printed some buttons for them, and then flew to Shenzhen to work on Core 2 Duo with Eric!

Raw bead-blasted watches.

Peanut does IQC.

Anodized blue watch cases.

Bobby tells me how to get to the hotel.

I wore this for a few months! Then I killed it from water incursion. Also, the antenna started acting funny at some point.

9 Likes

In August, I built up another one. I spent a lot of time polishing a watch. Titanium is a huge pain in the ass to polish: it work-hardens, and it doesn’t conduct heat away very well, so it means you absolutely nuke your abrasive from heat build up. The results still had a bunch of superficial scratches, but it was not bad for a first try. I anodized on a low-voltage fade.

A watch brought up to a polish. This took me forever.

Shiny watch on my X2100’s keyboard.

Fresh out of the ano bath.

Pressing buttons in.

Ready for assembly!

On my wrist, out in the sun :heart:

This was a great watch. I assembled this with some buttons I got from a UK supplier, Cousins. (Unfortunately, they were a completely insane $7 each…)

I daily drove this watch for a a few months. Eventually, it had the same fate as the previous one – it eventually ‘became deaf’, and would only talk to my phone if it was a few inches away. I thought the problem on the previous one was that water incursion had killed the antenna, but this was a second one!

I did some looking, and the root cause was even more subtle. This screw did not seat well into the chassis (we had a place to glue a FDM ‘nut’ in there, but it didn’t really compress the PCBA into placed):

It turns out that this screw is vital to make the antenna work. It presses the PCBA against that little gold thing on the screen assembly, which is the contact for the antenna. If it doesn’t press hard enough, there forms an air gap… then the watch’s RF performance suddenly gets a lot worse.

We learned a lot from the first gen design. Now it’s time to do it again – ideally, this time with actual threads in the material.

I put a Snowy back on my wrist, and forgot about it for a few months.

3 Likes

Ooh that anodised aluminum case looks super pretty

1 Like

We did another run a few months ago. @astosia made some design changes so that the internal stamped sheet metal snapped into place better, and we asked ProtoTi to tap the threads to be M1.2. (I sure did NOT have any interest in tapping M1.2 threads in titanium myself.) Inexplicably, I apparently didn’t bother to take any photos, but they arrived on December 1st.

But the real excitement was that I found a vendor on Alibaba that would make a reasonable quantity of buttons for us! We got some custom press-in buttons to our specification. On December 29th, a reasonable quantity of custom buttons arrived.

You know, a reasonable quantity. Like… 200 of them.

The buttons didn’t quite press in. It turns out that you need to specify a press fit worth of tolerances if you want a press fit. You learn something new every day! So I bought some reamers, and off I went to take the shells over to my friend Craig’s house to use his drill press. His shop cat, Mina, came to help out.

I processed all the holes to make sure they were good and round:

And at last, I had holes that the buttons fit into. Unfortunately, they fit… with a clearance fit. Grr. I ordered some Loctite 680, which is a particular retaining compound exactly designed to hold metals in place for clearance fits.

Of course, since the buttons were steel, I couldn’t install the buttons until all the titanium had been anodized. I wanted to try doing some mask work this time, and I wanted to experiment with various other surface finishes – I wanted to do a brushed finish, and I was hoping for some high voltage pinks. Well, the masking didn’t go so well, and neither did the high voltage finish! I didn’t get the sharp lines I wanted, and the paintbrush I used to paint on electrolyte also plated some other junk metal onto the part, and also, I had some contamination so no matter how much current I stuffed into it, I was definitely not going to get up to those 70V colors. It looked like I was going to have to actually get serious about etching contaminants away in advance. In for a penny, in for a pound – I ordered a bottle of Multi-Etch.

I grumbled, and started to polish off the previous attempt at an anodization. I did get one reward for it, though: since I put a brushed finish on the part before, when I lightly polished, I first polished the tops of the ridges, and left the anodization deep into the brush marks! The result is that I got a very interesting anisotropic finish – if you look at it straight on, you get color, but off angles, you get the shine from the tops. The process was not well controlled in this case, but you could see what I was going for, and what sort of effects were possible – some very angle-sensitive pastels.

I grumbled at this project having taken over my kitchen table for much longer than I had wanted, and waited for my bottle of Multi-Etch to arrive in the mail.

2 Likes

Just to let you know, I enjoy reading about your adventures. Brings back fond memories of working in an engineering lab!

2 Likes

Another me-to reply, I loved reading about the process you went through. I hope to vicariously read some more :wink:. Good luck with the project!

Those cases look awesome.

2 Likes

All the goodies arrived recently! I had hoped to take a crack at it today. That didn’t happen, for reasons involving taking a nap on the couch. But maybe tomorrow…

3 Likes

Friends, the frustration is real. Since we last spoke, all hell has broken loose.

I finally got a chance a few weeks ago to unbottle the Multi-Etch, and mix up some TSP, and pour it into tupperwares, and take a crack at anodizing another Pebble. I decided I wanted to do a nice high-voltage gradient, from about 50v to 80v. I was feeling good about the prospect, so I grabbed my brand new Insta360-brand GoPro, and stuck it on a chest mount, and began recording.

The etch was supposed to be slow at room temperature, but man, it was really slow. I gave it a quick etch, which didn’t seem to do much, so I gave it a heavier etch. Then, I put it in the TSP tank. I got a light coat on it, but anodization stopped at about 5V. Not good. I went all around the thing, and managed to get an uneven anodization if I dunked some corners in, but I couldn’t reliably anodize the whole thing. There were definitely metal inclusions.

I took a crack at blasting out the metal inclusions and trying to reduce them directly – I turned the power supply around, and cranked the current up. The only thing I got from that was… bubbles. Lots of bubbles. I flipped it around one more time and turned the current up in the hopes of getting some colors around the little inclusions, and… well, they stuck out pretty seriously.

The pitting was really deep. It seemed as much as like 100um deep. This didn’t seem like something I could have screwed up myself. As far as I can tell, I think this is a contaminant in the sintering process – maybe some steel particles that got in with the titanium.

Discouraged, I was curious if the etching even did anything at all for contaminants. I tried something that I expected to bake in some pretty serious contaminants – I hit the part with a steel wire wheel to try to put a bit of a brushed surface finish onto it. Then, I put it in the multi-etch with a heating pad under it, and let it sit for a few minutes. It did a very nice job of hazing the surface, which indicates to me that the etching process was doing something. Did it take down the steel? I took it upstairs, cranked the anodization current up to the moon, and gave it a shot.

The answer is: no! Apparently multi-etch really does not do anything against taking steel surface contaminants out, and instead just pulled oxides off. Kind of a disappointment, but also, it’s good to know what it’s capable of!

Having one scrap unit at my desk (well, ok, maybe this was the “starry night” surface finish?), I grabbed the new bottle of Loctite 680 I had at my desk, and glued in the buttons.


If nothing else, that part came out ok. The new buttons fit, and are retained well by the compound!

4 Likes

Looks cool, even if it wasn’t what you were aiming for!

2 Likes

The election results are tallied and ratified! It is up to the next Board to tell you what they are, but I can tell you that I’m not on it, and now I get to work on my own projects with my Pebble time (Pebble Time time?).

To celebrate, here’s a titanium Pebble with a brushed surface finish.

Then, I went for a bike ride to the office. It happened to be the golden hour.

Tomorrow, some AliExpress diamond polishing paste arrives.

5 Likes